summaryrefslogtreecommitdiffstats
path: root/src/intel/compiler
Commit message (Expand)AuthorAgeFilesLines
* anv,radv,turnip: Lower TG4 offsets with nir_lower_texJason Ekstrand2019-03-211-0/+1
* intel/nir: Lower array-deref-of-vector UBO and SSBO loadsJason Ekstrand2019-03-151-0/+11
* i965: Stop setting LowerBuferInterfaceBlocksJason Ekstrand2019-03-151-1/+0
* i965: Disable ARB_fragment_shader_interlock for platforms prior to GEN9Plamena Manolova2019-03-141-0/+1
* intel/nir: Combine store_derefs to improve code from SPIR-VCaio Marcelo de Oliveira Filho2019-03-131-0/+1
* intel/nir: Combine store_derefs after vectorizing IOCaio Marcelo de Oliveira Filho2019-03-131-0/+1
* nir: Add a pass to combine store_derefs to same vectorCaio Marcelo de Oliveira Filho2019-03-131-0/+1
* intel/fs: Fix opt_peephole_csel to not throw away saturates.Kenneth Graunke2019-03-121-0/+1
* intel/nir: Vectorize all IOJason Ekstrand2019-03-121-0/+17
* intel/nir: Move lower_mem_access_bit_sizes to postprocess_nirJason Ekstrand2019-03-081-2/+1
* intel/compiler: silence unitialized variable warning in opt_vector_float()Brian Paul2019-03-081-1/+1
* intel/nir: Move 64-bit lowering laterJason Ekstrand2019-03-061-21/+13
* nir/lower_doubles: Inline functions directly in lower_doublesJason Ekstrand2019-03-062-19/+6
* intel/nir: Drop an unneeded lower_constant_initializers callJason Ekstrand2019-03-061-2/+0
* intel/debug: Add a debug flag to force software fp64Jason Ekstrand2019-03-061-1/+1
* intel/fs: Fix extract_u8 of an odd byte from a 64-bit integerIan Romanick2019-03-061-0/+7
* intel/fs: nir_op_extract_i8 extracts a byte, not a wordIan Romanick2019-03-061-2/+4
* intel/compiler: Silence unused parameter warning in brw_interpolation_map.cIan Romanick2019-03-063-7/+4
* intel/compiler: Silence many unused parameter warnings in brw_eu.hIan Romanick2019-03-061-9/+9
* glsl: rename is_record() -> is_struct()Timothy Arceri2019-03-061-2/+2
* spirv: Use the same types for resource indices as pointersJason Ekstrand2019-03-052-0/+2
* intel,nir: Lower TXD with min_lod when the sampler index is not < 16Jason Ekstrand2019-03-041-1/+3
* nir/glsl: Add another way of doing lower_imul64 for gen8+Sagar Ghuge2019-03-042-0/+12
* intel/compiler: Move int64/doubles lowering optionsJordan Justen2019-03-022-34/+39
* intel/fs: Don't assert on b2f with a saturate modifierIan Romanick2019-03-021-1/+3
* intel/compiler: Add commas on final values of compaction table arraysMatt Turner2019-03-011-15/+15
* intel/fs: Generate if instructions with inverted conditionsIan Romanick2019-03-011-3/+20
* intel/fs: Emit better code for b2f(inot(a)) and b2i(inot(a))Ian Romanick2019-03-012-0/+40
* intel/fs: Use De Morgan's laws to avoid logical-not of a logic result on Gen8+Ian Romanick2019-03-011-0/+59
* intel/fs: Emit logical-not of operands on Gen8+Ian Romanick2019-03-012-6/+27
* intel/fs: Refactor ALU source and destination handling to a separate functionIan Romanick2019-03-012-33/+60
* intel/fs: Handle OR source modifiers in algebraic optimizationIan Romanick2019-03-011-1/+10
* intel/fs: Relax type matching rules in cmod propagation from MOV instructionsIan Romanick2019-03-011-4/+19
* intel/vec4: Emit constants for some ALU sources as immediate valuesIan Romanick2019-03-011-0/+71
* intel/compiler: Re-prefix non-logical surface opcodes with VEC4Jason Ekstrand2019-02-2810-47/+25
* intel/schedule_instructions: Move some commentsJason Ekstrand2019-02-281-37/+37
* intel/compiler: Drop unused surface opcodesJason Ekstrand2019-02-287-60/+0
* intel/fs: Get rid of the IMAGE_SIZE opcodeJason Ekstrand2019-02-283-10/+5
* intel/vec4: Drop dead code for handling typed surface messagesJason Ekstrand2019-02-285-272/+0
* intel/fs: Drop the fs_surface_builderJason Ekstrand2019-02-284-469/+195
* intel/fs: Re-order logical surface argumentsJason Ekstrand2019-02-281-2/+2
* intel/fs: Add an enum type for logical sampler inst sourcesJason Ekstrand2019-02-283-39/+58
* intel/compiler: use correct swizzle for replacementLionel Landwerlin2019-02-271-5/+5
* i965: Add support for sampling from XYUV imagesKasireddy, Vivek2019-02-262-0/+2
* intel/eu: Add an EOT parameter to send_indirect_[split]_messageJason Ekstrand2019-02-254-19/+29
* intel/fs: Rely on undocumented unrestricted regioning for 32x16-bit integer m...Francisco Jerez2019-02-211-3/+11
* intel/fs: Implement extended strides greater than 4 for IR source regions.Francisco Jerez2019-02-211-3/+10
* intel/fs: Cap dst-aligned region stride to maximum representable hstride value.Francisco Jerez2019-02-211-5/+23
* intel/fs: Lower integer multiply correctly when destination stride equals 4.Francisco Jerez2019-02-211-4/+8
* intel/fs: Exclude control sources from execution type and region alignment ca...Francisco Jerez2019-02-213-4/+68
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-25 23:18:47 +0000